摘要
采用当前方法修复电力物联网电能质量数据时,修复数据花费的开销较大,修复数据所用的时间较长,存在修复通信开销大和修复效率低的问题。提出基于Shapelets的电力物联网电能质量数据修复算法,根据电力物联网各节点获得的监测数据构建电能质量时间序列,通过关键趋势转折点的分段线性对时间序列的时间特征和趋势特征进行表示,建立时间序列模式特征矩阵。利用时间序列模式特征矩阵构建电力物联网电能质量数据修复模型,采用差分进化算法对电力物联网电能质量数据修复模型进行求解,实现电力物联网电能质量数据的修复。仿真结果表明,所提方法的修复通信开销小、修复效率高。
At present,the cost of repairing the data in electric Internet of Things is large,and the time to repair the data is too long,leading to large communication overhead and low repair efficiency.Therefore,an algorithm based on Shapelets to repair the power quality data in electric IoT was proposed.The time series of power quality was constructed through the monitoring data obtained by the nodes in electric IoT.The time feature and trend characteris-tic of time series were represented by the piecewise linearity of key turning points.And then,the time series pattern feature matrix was constructed.On this basis,a model to repair power quality data was built.Finally,the differential evolution algorithm was used to solve the model of power quality data repair and thus to realize the restoration of pow-er quality data in electric IoT.Simulation results prove that the proposed method has small communication cost and high repair efficiency.
作者
王跃晟
王维庆
WANG Yue-sheng;WANG Wei-qing(Engineering Research Center for Renewable Energy Power Generation&Grid-connected Technology of Ministry of Education,School of Electrical Engineering,Xinjiang University,Xinjiang Urumqi 830047,China)
出处
《计算机仿真》
北大核心
2020年第12期85-89,共5页
Computer Simulation
基金
国家自然科学基金(51667020)。
关键词
时间序列
电力物联网
数据修复
Time series
Electric Internet of Things(EIOT)
Data repair